Veo 3 introduces native audio generation for AI videos, pushing beyond the silent clips of earlier models. It produces eight-second, 720p videos from text prompts, complete with realistic physics, expressive motion, and lip-synced speech. Developers can now access it in paid preview mode through the Gemini API and Google Cloud’s Vertex AI platform. Pricing starts at $0.75 per second for video with audio, with a more affordable “Veo 3 Fast” version expected soon.

The launch positions Google as a contender in multi-modal AI content generation, targeting creative developers, game studios, and production teams. Early partners like Cartwheel and Volley are using Veo 3 to build 3D animation tools and in-game cut-scenes. This move signals Google's effort to open new developer workflows for generative media while strengthening its ecosystem against competitors like OpenAI’s Sora and Runway.